用语音活动时间特征实时区分电话留言与真人应答，准确率达96.1%。
Real-Time Voicemail Detection in Telephony Audio Using Temporal Speech Activity Features
- 从预训练语音检测器提取15个时间特征，用轻量树模型分类。
- 在764段录音上达96.1%准确率，生产环境误报仅0.3%。
- 无需GPU，46毫秒完成推理，适合高并发通话系统部署。
外呼AI系统需实时区分语音信箱问候与真人应答，以避免资源浪费和通话中断。本文提出一种轻量级方法：从预训练神经语音活动检测器（VAD）中提取15个语音活动的时间特征，再通过浅层树集成模型进行分类。在总计764段电话录音的两个评估集上，系统整体准确率达96.1%（734/764），其中专家标注测试集为99.3%（139/140），保留生产集为95.4%（595/624）。在7.7万次真实通话的验证中，保持0.3%误报率与1.3%漏报率。端到端推理仅需46毫秒（双核通用CPU，无GPU），支持380+并发WebSocket通话。在3,780种模型、特征与阈值组合的搜索中，特征重要性集中于三个时间变量。加入语音转写关键词或蜂鸣音特征未提升最佳实时配置性能，且显著增加延迟。结果表明，语音活动的时间模式是区分语音信箱与真人应答的强信号。
原文摘要 · Abstract (English)
Outbound AI calling systems must distinguish voicemail greetings from live human answers in real time to avoid wasted agent interactions and dropped calls. We present a lightweight approach that extracts 15 temporal features from the speech activity pattern of a pre-trained neural voice activity detector (VAD), then classifies with a shallow tree-based ensemble. Across two evaluation sets totaling 764 telephony recordings, the system achieves a combined 96.1% accuracy (734/764), with 99.3% (139/140) on an expert-labeled test set and 95.4% (595/624) on a held-out production set. In production validation over 77,000 calls, it maintained a 0.3% false positive rate and 1.3% false negative rate. End-to-end inference completes in 46 ms on a commodity dual-core CPU with no GPU, supporting 380+ concurrent WebSocket calls. In our search over 3,780 model, feature, and threshold combinations, feature importance was concentrated in three temporal variables. Adding transcription keywords or beep-based features did not improve the best real-time configuration and increased latency substantially. Our results suggest that temporal speech patterns are a strong signal for distinguishing voicemail greetings from live human answers.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。